Although epidemiological studies suggest that there is a weak relationship between exposure to electromagnetic fields (EMF) and cancer there are at present few plausible molecular mechanisms to explain this effect. The known involvement of free radicals in the etiology of cancer and other diseases raises the possibility that EMF may somehow act by increasing the lifetime and/or concentration of free radicals. It has been known for many years that in organized media (e.g. detergent micelles) applied magnetic fields cause an increase in average free radical concentration and lengthen the lifetime of free radicals. If these effects occur in vivo then this may explain some of the reported biological effects of EMF. Our initial studies have focused on the intact erythrocyte as a test system that could be used to study the effects of MF on free radical processes. We have found that ketoprofen, a phototoxic, non-steroidal anti-inflammatory drug (structurally related to benzophenone), caused complete hemolysis of human erythrocytes after only 20 min UV irradiation. Photohemolysis was dependent on the presence of oxygen and was inhibited by reduced glutathione and butylated hydroxytoluene (BHT) suggesting that the mechanism involved lipid peroxidation. When UV irradiation was carried out in a magnetic field (3500G) the time taken for hemolysis to occur in 50% of the cells (t+) was significantly shortened (78 min vs. 96 min). A significant decrease in t1/2 could be observed at fields as low as 250G but the effect disappeared at 100G. These observations can be attributed to a magnetic field induced decrease in the rate of intersystem crossing of the geminate triplet radical pair generated by the reduction of ketoprofen in its triplet excited state by erythrocyte membrane components, probably lipids. To our knowledge this is the first demonstration of an effect of MF on a toxicological process that is known to occur via a free radical mechanism.